FirstWatt J2

Manniraj,

Those are nicely sized heatsinks - what is the source / manufacturer?

Sorry did not check this thread for a long time 🙂

I got these heat sinks locally in India and most of my all class A amps use the same sized and good enough.

By the way I decided to go dual mono for J2 as well and got my 18v toroidals pair and going to use pair of the universal CRC psu boards already populated using 12000uf caps.
 
Sorry did not check this thread for a long time 🙂

I got these heat sinks locally in India and most of my all class A amps use the same sized and good enough.

By the way I decided to go dual mono for J2 as well and got my 18v toroidals pair and going to use pair of the universal CRC psu boards already populated using 12000uf caps.


I only have the dual version of this amp, but it's one of may favs. I liked it so much I bought an extra pair along with the unobtainable parts. Every now and then I peak in the draw and see them there glowing invitingly. 😀
 
After 8 years trying to make nearly perfect J2......

You see in the first picture the state in the last years, h2 over h3 is o.k. with around 12-14dB higher.
But since the program Diana is published and we can see the residual as well, I saw that I have h2 with positive phase, second picture.

After some fumbling today with the source and drain resistors between the Semisouths I finally got the negative phase....what a moment!
Picture three.

The four SS were bought some time ago from Semisouthfan, as quartets. It seems they had greater differences in transconductance than I thought.

Now I must elaborate how the OLG changed, because the sound depends a good deal how much loop gain (difference of OLG and CLG) exists.
Maybe I have to change the feedback resistor and the CLG value.

I can only recommend to check your builds with Diana.

Hi Generg! I was looking back and remembered this post about your discovery of the positive phase 2nd, and how you tweaked it to negative. I noticed in the REW screenshot you attached of the before (positive 2nd), the phase angle next to the distortion measurement indicates a negative. I’ve never quite trusted the phase angles that REW provides (assume its relative to fundamental?), but I’m curious if you’ve found the REW phase to typically represent 180deg opposite the phase you observe with Diana? Though I currently don’t have the means to view the distortion residual in my amps, listening tests and lots of back and forth with the speaker leads have led me to think this is the case.
 
I had the idea also if the phase connotation of the REW and that Diana gives have a fixed correspondence. I did not prove it systematically. There was a comment from someone that the value REW or Spice shows might depend if the amp switches itself the phase by its circuit.
I will have a look at it. I have only one FW case and in the moment the M2 is inside.
A gift from genius ZM.
When I go back to the J2 I will look for this!

:--))
 
I have been asked various questions about alternative devices for the J2.
If anyone, Nelson who designed the circuit should be the one to answer those questions, not me.
I just mess around for my own interest.

The sonic signature of the J2 circuit is largely determined by the front end diff pair and the bottom power device.
If you replace those by something else, you will change the character and should not call it J2 anymore.
The modulated current source at the top rail has less influence and one can consider alternatives there.
I have already explained why I prefer 2SK3497, but that is my prejudice for Japanese FETs.
Nelson has mentioned IRFP240, others IRFP150 or IRFP250.
They will all work, but you might need to adjust the power resistor values at the output to adjust the level of current modulation.
The values shown in the 3497 schematics are default values which gives equal dynamic current sharing between the Semisouth and the Toshiba.
Individual FETs with different transconductances will change the picture somewhat.
That is the only reason why you want to use matched pairs for stereo, so that the 2 channels are identical.

Can you use one of those MOSFETs to replace the Semisouth, functionally ?
The simple answer is no.

The complex answer is yes, if you know what you are doing.
All these high Yfs MOSFETs have large positive tempco, where Semisouth has close to zero.
Also the Vgs at bias is hugely different.
So you need to change the darin resistor value, and maybe use a NTC in parallel.

The J2 is a costly project.
If you are not experienced in DIY and do not understand circuits, then my advice is DO NOT attempt this.
Any circuits that look simple have hidden complications that can make things go wrong.
It is unfair to rely on anyone here to carry you along and solve your problems.
Try something suitable for beginners first, like Nelson's excellent ZEN series amps.

You have been warned.



Patrick
 
J2CLONATO doppio mos fet

hi I am an Italian autoscanner after having recently made F5 turbo with some satisfaction more than other projects previously made I am intrigued by the good comments regarding the final class A J2 clone. Reading the various posts I noticed that many enthusiasts are held back in the realization due to the very expensive JFET .... R100 and therefore replacing it with normal mos fet, consequently changing the characteristics of the amplifier. I would like to know, possibly citing the great N. Pass. what happens if I modify the J2 scheme in this way: For the CCS I would use a mos fet type FQA19N20 (22 amper) and while for the signal I would put in parallel n. 2 Mfet 2SK3497.
Now being little expert on some functions of mos fet and not having professional equipment I would like to know if I get benefits or I go to worsen the circuit formed with simple MosFets also type IRFP240
 

Attachments

  • clone J2 MODIFICATO.jpg
    clone J2 MODIFICATO.jpg
    223.3 KB · Views: 603
So doubling the Mosfet on the signal has no benefit and it is advisable to use the original scheme by changing to the limit as nelson pass says with IRFP240
Is this the new scheme of a good J2-like finish?
which mos fet would you recommend to me among the three I would like to use?
IRFP240 - IRFP044 - FQA19N20? maybe even with a different combination between CCS mosfet and Mosfet for the signal
 

Attachments

  • clone J2.jpg
    clone J2.jpg
    445.4 KB · Views: 779
Using 2x J174 in parallel will reduce noise by 1.4x, and increase Yfs a bit.
But it is still far from the original 2SJ74s.

On top of that, the Semisouth has a much lower bias voltage than any of those you named.
So you will have to change more things in the circuit, e.g. R1.

I suggest you do some simulations first and satisfy yourself that your schematics works.
That means the performance meets your expectation and it is stable under reactive load.

If you do not want to start designing, but just want to build without unobtanium parts, then maybe you should take a look at :
UDNeSS, or You don't need Semisouth's


Cheers,
Patrick
 
Hi all,
my SS are dying ... crackling / distraction sound sometimes so ich changed the "good" ones from the upper one to the lower place and put some IRFP150 upper side.
After a few week (10 hours a day) the same game. :-( I think they were counterfeits but they doesn't look so.
Does anyone know such behaviour?
 
Thanks,
when it occurs it is over over the entire spectrum, curve tracer says they are ok. :-(
And it is not always ... sometime around a few minutes and then the rest of the day is all ok.
I checked the iso pads an so, changed the few parts but all time it comes back. The only parts I left inside was the SS and the IRF part. 170/74er are also changed and they are the Linear parts from the store.

In short I a have 4 SS in stock I don't know their status. :-( Expensive try with the last two. :-(
 
Sorry for late response ...

Solder issue I think not because it is now the second pair SS (in lower position) and I changed all other parts and checked more then two times.

From a diyAudio forum member, he sold more SS here and I didn't hear from problems with his parts.
A curve trace looks like an original (before using them in my J2).